Introduction to Economic Evaluation Part IPart I
April 20, 2011
Webinar Content
• Overview of economic evaluation (EE) methods
• Cost of Illness (COI) analysis or economic impact analysisp y
• Programmatic cost analysis (CA) g y ( )
What is Economic Evaluation (EE)?Applied analytic methods to:
Identify,MeasureMeasure,Value, andCompare
th t d fthe costs and consequences of treatment* and prevention** strategies.
* Done a lot** Done “not so much”
Wh C Ab t EE?Why Care About EE?
Maximizing outcomes is important.
Minimizing costs is important too Minimizing costs is important too.
Resources are limited, so hard (resource allocation) decisions must be made.
Demonstrates the value provided from theDemonstrates the value provided from the resources expended (return on investment).
EE Methods
• Partial evaluation – costs only– Cost of illness (COI) analysis – Economic Impact
AnalysesAnalyses– Cost analysis - programmatic cost analysis (CA)
• Full evaluation – costs and outcomes– Cost-benefit analysis (CBA); Benefit-cost analysis
(BCA)(BCA)– Cost-utility analysis (CUA)– Cost-effectiveness analysis (CEA)– Cost-effectiveness analysis (CEA)
f C ( C )Benefit-Cost Analysis (BCA)St d di b th t d t i $• Standardizes both costs and outcomes in $
Includes health and nonhealth outcomes• Includes health and nonhealth outcomes
P id bj ti• Provides an objective summary measure• Net Benefits (B – C), Net present value (NPV)• Benefit-cost ratio (B / C)• Benefit-cost ratio (B / C)
Cost Utilit Anal sis CUACost-Utility Analysis - CUA• A method used to compare costs and benefits of
interventions where benefits are expressed as theinterventions where benefits are expressed as the number of life years saved adjusted to account for loss of quality.q y
• Combines• Length of life (survival), and• Quality of life
• Compares disparate outcomes in terms of utility• Quality adjusted life years (QALYs)• Quality-adjusted life years (QALYs)• Disability-adjusted life years (DALYs)
• Derives a ratio of cost per health outcomeDerives a ratio of cost per health outcome• $/QALY or $/DALY
Cost-Effectiveness Analysis (CEA)(CEA)
• Expresses outcomes in natural units.p• (e.g., number of cases prevented or lives saved)
• Compares results with other interventions ff ti th t
paffecting the same outcome.
Derives a ratio of cost per unit of outcome• Derives a ratio of cost per unit of outcome.• $/case prevented• $/life saved• $/life year saved$/life year saved
Cost-effective = Cost-saving
20000
25000
d
5000
10000
15000
/LY
sav
ed
-5000
0
5000
$/
0% 20% 40% 60% 80% 100%
% people vaccinated
Wh t EE M th d t U ?Ex: Health vs. Defense
B fit $
What EE Method to Use?Benefits = $
BCA
Ex: Violence vs ObesityEx: Violence vs. ObesityBenefits = QALYs
CUA
Ex: Child Maltx vs. Child Maltx
Benefits = CasesCEACEA
E i I tEconomic Impact or Cost of Illness (COI) Analysis
E ti t t t l t i d b f• Estimates total costs incurred because of a disease or condition.– Costs of medical resources to treat disease/injury.Costs of medical resources to treat disease/injury.– Costs of non-medical resources to treat disease/injury.– Losses in productivity due to disease.
• Uses:– To estimate economic burden of disease/injury– Benefits measure in BCA– In CUA, medical costs averted in numerator of CE ratio– In CEA, medical and productivity losses averted in
t f CE tinumerator of CE ratio
Medical/Non-medical Costs• Medical costs
• Inpatient• Outpatient• Outpatient• MH• Rx
• Non-medical costs• Legal and justice system• Travel expenses to seek treatment• Childcare while seeking treatment
Productivity Losses HumanProductivity Losses – Human Capital Approach
• Assumes– Value of a worker’s productivity=earnings, bc a profit-
maximizing employer will not pay a worker more than themaximizing employer will not pay a worker more than the additional value he/she contributes
• Valuation is simplest when an intervention affects mortalitymortality– Categorizes work loss (gain) by age, sex, occupation– Calculates the discounted value of expected labor– Uses gross earnings (before taxes) and includes fringe
b fitg g ( ) g
benefits– Imputes value for non-market labor such as household
productivity– Occasionally subtracts future consumption of goods andOccasionally subtracts future consumption of goods and
services
HC Approach - Valuation of MorbidityMorbidity
• Is more complicated than valuation of death• Valuation may not simply be the number of work• Valuation may not simply be the number of work
days lost, bc– Ability to return to work may be occupationally relatedy y p y– Returning to work doesn’t necessarily mean resuming
the same level of productivityThe change in health status may necessitate job– The change in health status may necessitate job switching
COI ReportingCOI Reporting• Prevalence-based costs
Cross sectional data– Cross-sectional data– Includes all costs within a specific time period,
regardless when event occurred– Useful for thinking about resources required for
treatment within a given time period
• Incidence-based, or lifetime, costs– Longitudinal data– Includes lifetime costs for new events– Useful for thinking about potential savings from
preventionprevention
COI Methods• All medical costs for the victimsAll medical costs for the victims
• Only diagnosis-specific costs for the victimsOnly diagnosis-specific costs for the victims– Add attributable fraction
• Incremental cost approach– Match against controlg– Regression– Attributable fraction
Sum of All Medical Costs• Prevalence-based, cross-sectional data
• Sum all annual medical costs comparing those reporting disease to those not reporting diseasedisease to those not reporting disease
• Pros: Good for relative comparisons and understanding of impact on the healthcare system
• Cons: Does not isolate the economic burden associated• Cons: Does not isolate the economic burden associated with the smoking, or understanding of co-morbidities
Sum only Diagnosis-Specific Costs• Pros:
– Represents lower-bound actual costs of disease– Good for incidence-based models
• Cons: M d ti t t if bid t t– May underestimate costs, if co-morbid events are not included.
Attributable Fraction• Also includes the indirect health expenditures
associated with the behavior, through other diseases or conditionsdiseases or conditions
The attributable fraction is added to the total• The attributable fraction is added to the total diagnosis-specific costs
Ruff, Volmer, Nowak, & MeyerEuropean Respiratory Journal
THE ECONOMIC IMPACT OF
European Respiratory Journal2000; 16: 385-390
THE ECONOMIC IMPACT OF SMOKING IN GERMANY
Methods• Sources of informationSources of information
– Official German statistics– MEDLINE– German Institute for Medical Documentation and
InformationI t t– Internet
– Health insurance companies databasesGerman Federal Institute of Statistics– German Federal Institute of Statistics
Methods• Direct CostsDirect Costs
– Healthcare resources: ambulatory care, drug treatment, hospital care, rehabilitation, long-term careR it lti li d b it t– Resource units multiplied by unit costs
• Indirect Costs– Costs incurred from productivity losses premature– Costs incurred from productivity losses, premature
retirement, and premature mortality– Human Capital Approach
• Life years lost before age 65 multiplied by gross annual income
• Intangible Costs (not included)Patient’s quality of life (no reliable nationwide data)– Patient s quality of life (no reliable nationwide data)
Methods
Results
Results
Common Mistakes in COI AnalysesU i t ti t f diff t• Using cost estimates from different sources without converting into the same base year.Not adjusting market prices to reflect true• Not adjusting market prices to reflect true costs.
• Combining reporting methods in same analysis• Combining reporting methods in same analysis– Annual vs lifetime costs
• Not using present value of future potential• Not using present value of future potential earnings.
So What?COI hi hli ht th it d f th b d• COI highlights the magnitude of the burden relative to other burdens.It id d t t f t• It provides data to argue for more resources to prevent the burden.
• Used with estimates of costs to prevent the burden, COI can provide policy-makers with return on investment informationreturn-on-investment information.
Programmatic Cost Analysis• Cost analysis is an economic evaluationCost analysis is an economic evaluation
technique that involves the systematic collection, categorization and analysis of program costs.
• Costs are the value of the resources (people, (p p ,buildings, equipment and supplies) used to produce a good or a service.
What are Resources?Resources are the:Resources are the:• People• Places and• Places, and • Things needed to carry out a task.R h t d !Resources are what we spend our money on!
What they cost can be computed many different waysdifferent ways.
C t A l iCost Analysis• Typically the first step in economic evaluation.
• Estimates total costs of running a program.
• Important for realizing costs from varyingImportant for realizing costs from varying perspectives. – e.g., incurred by program, incurred by participant
• Important for budget justification, decision making, and forecasting.
• Type of cost analysis warrants which costs are included.
IMPORTANT t f i t• IMPORTANT: not focusing on program outcomes. . . yet.
• Financial Costs
Financial CostsFinancial Costs– Monetary expenditures for resources required to
implement the program – based on market prices– Typically found in the budget proposal– Typically found in the budget proposal– A convenient, but sometimes incomplete, way to
measure costs
• Examples:– Salaries for project personnelj– Supply costs– Computer purchases– Cost of curriculum materialsCost of curriculum materials
Economic CostsEconomic Costs• Economic Costs
(O t it t) Th l f th f• (Or opportunity cost): The value of the forgone benefit because the resource is not available for its next best use. E i t th t ’ t i th• Economists argue that a resource’s cost is the sacrifice necessary to obtain goods or services.
E l• Examples:• Volunteer time• Donated space (e.g., from a University)p ( g , y)
•• Shadow prices Shadow prices used when market price does not accurately reflect the value of the good.accurately reflect the value of the good.
• Direct costs
More Ways to Talk About “Costs”Direct costs – Those costs that are directly related to the project or
intervention or the specific task at hand
• Indirect costs– Also called overhead costs – these are the “support”
costs, including admin, accounting/finance management office supplies maybe utilities etcmanagement, office supplies, maybe utilities, etc….
Cost Analysis Terms• Fixed Cost• Fixed Cost
– Cost incurred only one time regardless of the number of times the intervention is conducted, i.e., the cost of staff training
• Variable Cost– Cost that varies with the the number of times the
intervention is conducted or with the number of clients covered i e the amount of condoms perclients covered, i.e., the amount of condoms per clients.
34
Total Costs (TC)
where P1 = Value of resource 1Q1 = Quantity of resource 1 usedP2 = Value of resource 2Q2 = Quantity of resource 2 used…Pn = Value of resource nQn = Quantity of resource n used
Average Costs (AC)
• Total resource costs divided by total units of output.
AC = TC / Q
WhereTC Total costsTC = Total costsQ = Total units of output
• The marginal cost is the resource costs Marginal Cost (MC)g
associated with producing one additional (or one less) unit of the same intervention.
MC = (TC’ – TC) / (Q’ – Q)WhereWhere
TC = total costs at the current activity levelTC’= total costs at the higher activity level
Q = current output levelQ = current output levelQ’= higher output level
Average vs. Marginal Cost• The marginal cost can be lower or higher thanThe marginal cost can be lower or higher than
the average cost, depending on whether or not the program is operating at full capacity.
• When MC<AC, “economies of scale” are being , grealized.– This is the fixed versus variable costs issue
Additional Factors and Issues • ScalingScaling
– How many times you do the intervention, or how many clients you serve, will affect your average cost – be careful!
• Comparability– Must count costs the same way for a valid
comparisonT f bilit• Transferability– Prices differ from market to market, and
across timeacross time
Ritzwoller, Sukhanova, Gaglio & GlasgowAnnals of Behavioral Medicine
COSTING BEHAVIORAL
Annals of Behavioral Medicine2009; 37(2): 218-227
INTERVENTIONS: A PRACTICAL GUIDE TO ENHANCE TRANSLATION
Smoking Less, Living More Program• Targeted adult smokers scheduled for surgery orTargeted adult smokers scheduled for surgery or
diagnostic procedure• Randomized to treatment or control arm of studyy• 6-month intervention with telephone counseling,
newsletters, and health education,• Completed a cost analysis of the intervention
Five-Step Process1. Perspective of the Analysis1. Perspective of the Analysis2. Identify Intervention Components3 Capture Intervention Costs3. Capture Intervention Costs4. Data Analysis5 Sensitivity Analysis5. Sensitivity Analysis
Perspective of the Analysis• Must consider decision makerMust consider decision maker
– Medicare/Medicaid– Health planp– Community– Public health entity
• For future dissemination purposes include– Capital equipment– Prescription drugs– Technology used to deliver intervention
Identify Intervention Components• Research
– Removed from cost analysis– Examples
• Grant administration, IRB approval, assessments and testing (not part of intervention), manuscript preparation
• Development– Necessary to differentiate those that would need to be replicated in
the future and those that would notExamples– Examples
• Development of protocols and assessment, website design, telephone script production
• Implementation/Intervention– Recruitment
• Include costs present in replication but exclude research specific such as informed consent
– All Others• Labor, counseling, monitoring, supply, printing, and mailing costs
Capture Intervention Costs• Prospective collectionProspective collection• EXCEL-based templates collected on a monthly
basis• Personnel
– Staff filled in time logs chronicling by hours or FTEg g y• Staff estimated to have spent 5-10 minutes per
month recording resource use
Data Analysis• Staff not asked to categorize resource use asStaff not asked to categorize resource use as
research/development/implementation– Rather, researchers divvied the aggregated costs
based on description of activity and knowledge of the intervention program
Personnel time valued using salaries and benefit• Personnel time valued using salaries and benefit data of actual personnel
• Summarized as total intervention costs and cost• Summarized as total intervention costs and cost per intervention participant
Scale Analysis• Used fixed and variable costs to estimate howUsed fixed and variable costs to estimate how
the total cost would vary depending upon the number of enrolled participants
Conclusion• Cost analysis is an important first step in y p p
economic evaluation.
• Despite variation in data collected across sites• Despite variation in data collected across sites, cost data provides important preliminary information on how much it may cost to replicate th t h dthe program at hand.
• and more information about the relationship• and more information about the relationship between outcomes and costs for future EEs.
Content of Part II Webinar
• Overview of BCA and CEA
M i QALY f i CUA• Measuring outcomes, QALYs, for use in CUAs
Thank You!
Top Related